In many data storage networks, storage nodes may be cooperating to provide high availability or other storage services for the storage network. As one particular example, the storage nodes may be a high availability (HA) pair of storage servers.
With these high availability (HA) pair configurations data storage networks often require one or more dedicated spare disks on each of the storage nodes even though both storage nodes in the HA pair configuration have connectivity to all disk(s). The number of spare disks required for each of the storage nodes typically depends on factors, such as media type, RPM, checksum style, and disk size.
Unfortunately, a dedicated spare disk available at one storage node in a HA pair configuration currently cannot be used for failed disk replacement by the other storage node unless manually reassigned. As a result, the unavailability of a spare disk at the storage node when needed leads to an increased duration of the degradation window. Additionally, the unavailability of a spare disk at the storage node when needed negatively impacts read performance in the HA pair configuration and makes the HA pair configuration less resilient.